IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
Mapping double and single crop paddy rice with
Sentinel-1A at varying spatial scales and
polarizations in Hanoi, Vietnam
Kristofer Lasko, Krishna Prasad Vadrevu, Vinh Tuan Tran, Christopher Justice
 Delta such as the Hanoi Capital Region, where population
Abstract— Paddy Rice is the prevalent land cover in the
mosaicked landscape of the Hanoi Capital Region, Vietnam. In
this study, we map double and single crop rice in Hanoi using a
random forest algorithm and a time-series of Sentinel-1 SAR
imagery at 10m and 20m resolution using VV-only, VH-only, and
both polarizations. We compare spatial and areal variation and
quantify input band importance, estimate crop growth stages,
estimate rice field/collective metrics using Fragstats with image
segmentation, and highlight the importance of the results for land
use and land cover. Results suggest double crop rice ranged from
208,000 ha to 220,000 ha with 20m resolution imagery accounting
for the most area in all polarizations. Based on accuracy
assessment, we found 10m data for VV/VH to have highest overall
accuracy (93.5%, ±1.33%), while VV at 10m and 20m had lowest
overall accuracies (90.9%, ±1.57; 91.0%, ±2.75). Mean decrease in
accuracy suggests for all but VV at 10m, data from harvest and
flooding stages are most critical for classification. Results suggest
20m data for both VV and VH overestimates rice land cover,
however 20m data may be indicative of rice land use. Analysis of
growing season suggests average estimated length of 93–104 days
for each season. Commune-level results suggest up to 20%
coefficient of variation between VV10m and VH10m with
significant spatial variation in rice area. Landscape metrics show
rice fields are typically planted in groups of 3-4 fields with over
796,000 collectives and 2.69 million fields estimated in the study
area.
Index Terms—Radar applications, image classification,
agriculture, rice, land cover, spatial analysis, Vietnam, Red River
Delta.
I. INTRODUCTION
R ICE (Oryza sativa) is the staple crop for economic and
cultural livelihood throughout much of Southeast Asia,
including Vietnam. Production of paddy rice in Vietnam has
been expanding consistently over time with 32.5 million metric
tons produced in year 2000, to 45.2 million metric tons in the
year 2015 [1]. All the while, area under cultivation has only
slightly increased with 7.67 million ha in 2000 and 7.83 million
ha in 2015, suggesting notable agricultural intensification.
Intensification is critical especially in areas of the Red River
Manuscript received August 2, 2017; revised October 14, 2017 and
November 27, 2017; accepted December 13, 2017. This work was supported by
University of Maryland and American Society for Engineering Education
SMART scholarship.
K. Lasko and C. Justice are with the Department of Geographical Sciences,
University of Maryland, College Park, MD 20740, USA.
Copyright 2017 IEEE. Author submitted final version.
1
pressure and a robust economy are driving peri-urban
expansion into agricultural areas [2]. Thus, reducing the amount
of area under cultivation in some areas.
Accordingly, it is of increasing importance to develop
efficient methods for mapping paddy rice in Vietnam. The two
dominant rice-producing hubs in Vietnam are within the Red
River Delta and the Mekong River Delta, together accounting
for the majority of national rice production. Optical imagery
such as from MODIS or Landsat, are frequently cloud
contaminated during key agricultural stages (i.e. planting and
harvest), thus hampering mapping efforts [3]. C-band Sentinel-
1 imagery is capable of penetrating through cloud cover making
it the obvious choice for timely mapping of paddy rice.
Efforts to map paddy rice using SAR originated in the 1990’s
with selected images from ERS-1 C-band imagery with
approximately 12.5m-30m spatial resolution and a repeat pass
of 35-days [4]-[9], and also using C-band RADARSAT at
spatial resolution of 10m-100m and a repeat pass of 24 days
[10]-[12]. These pioneering studies were often stymied by lack
of quality ground-truth imagery, single polarization, or were
limited to small spatial scale studies due to high data volumes.
Subsequent studies began focusing on utilizing multitemporal
SAR over larger land areas for improved rice mapping and
testing various algorithms with ERS-1, ERS-2, and
RADARSAT [13]-[17]. Improvements in sensors and
availability of ground-truth data led to further more extensive
rice mapping studies using L-band SAR such as
ALOS/PALSAR or JERS-1 [18], [19], ENVISAT ASAR [20]-
[23], TerraSAR-X and COSMO-SkyMed [24]-[27]. Recent
studies include RADARSAT-2 data, object-oriented crop
mapping [28], [29], combined optical and SAR data [30]-[32],
and Sentinel-1 C-band SAR [33]-[35]. More details are
available in a recent review [36].
While a variety of strong classification methods exist, a
number of recent studies have employed random forests for
SAR applications, including wetlands mapping [37]-[39],
general agriculture/land cover [40], and in paddy rice mapping
with promising results [41]-[43]. Recent remote sensing studies
K.P. Vadrevu is with the Earth Science Office, NASA Marshall Space Flight
Center, Huntsville, AL, 35805 USA.
V.T. Tran is with the Faculty of Information Technology, Hanoi Pedagogical
University 2, Hanoi, Vietnam.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
have also included use of texture-based classification such as
Gray Level Co-occurrence Matrix (GLCM) with promising
results [22], [44], [45]. In addition to mapping paddy rice areal
extent, other studies have employed SAR for agricultural
phenology estimation with multi-temporal imagery [46], [47],
including studies with rice paddy fields and polarization
variation, [48]-[55]. While some of these studies applied time-
series filters to derive accurate phenology metrics, these are not
always requisite due to the unique dynamic range of paddy rice
signal from SAR [41]. The SAR signal seen in a time-series can
accurately capture different stages of crop growth useful for
monitoring biophysical variables as well as length of season,
planting and harvest dates. While local farmers and
stakeholders have knowledge of the crop phenology, SAR-
based information is useful for broader national or regional
operational monitoring of crops (i.e. crop conditions, status, or
health) which is a major goal of international initiatives such as
Asia-Rice Crop Estimation and monitoring (Asia-RiCE) project
or the Global Agricultural Monitoring (GEOGLAM) initiative
[56], [57]. The SAR observations can survey the entire region,
whereas optical data can often be obstructed by cloud. Thus,
SAR data can be useful for crop mapping and monitoring at
regular intervals.
II. STUDY AREA AND DATASETS
A. Study Area
The mosaic landscape of the Hanoi Capital Region includes
a variety of land cover types dominated by rice agriculture, as
well as other small-holder croplands, urban areas, small
plantations, and aquaculture. The typical rice field size in the
area is about 800m2 with fields routinely planted in large
collectives as necessary to facilitate irrigation. The field size is
suitable for moderate-to-fine resolution mapping [58]. The
Hanoi Capital Region is situated within the Red River Delta,
Vietnam’s oldest rice producing region, accounting for about
15% of the country’s total rice production [1]. Of all the
different crops, rice is most prevalent and it is the dominant
crop type in the region. It has two distinct seasons: Winter-
Spring, and Spring-Summer. Rice has three distinct stages:
sowing/transplanting, growth, and harvest/post-harvest, all of
which can be identified using satellite data [8]. In the study
area, rice is sown or transplanted after the Tet holiday in
February or early-March. Subsequently, a significant green-up
is observed as the rice matures in its vegetative stage,
especially after heading in April [59], [60]. During late-May to
June the rice is harvested, and rapidly prepared for the next
season starting in late-June or July. After the harvest in each
season (June and October) the rice residues including straw
and stubble, are regularly burned accounting for as much as
13% of PM2.5 emissions for Vietnam; thus burning may
constitute a significant air quality issue [58], [61].
The rice fields are supported by a vast network of irrigation
and drainage canals and small access roads seen throughout
the region. While most fields practice double cropping, some
fields cultivate a single crop of rice. After the harvest, single
rice fields remain flooded to support the growing aquaculture
industry [62], [63]. The study area has seen rapid expansion of
aquaculture with 16.5 thousand ha in year 2000, and 46
Copyright 2017 IEEE. Author submitted final version.
2
thousand ha in year 2014, with the vast majority of growth
occurring in Hanoi province [1].
1) Satellite data
The Sentinel-1 satellite from the European Space Agency
provides C-band SAR imagery (5.4GHz) near-globally with a
12 day revisit time or 6 day revisit time depending on
availability of Sentinel-1B imagery. The Sentinel-1 imagery is
provided as dual-polarized Interferometric Wide swath (IW)
data with vertical-transmit, vertical receive (VV), and vertical
transmit, horizontal receive (VH) polarizations. Each
polarization is at a nominal spatial resolution of 5x20m prior
to pre-processing using an open-access operational baseline
observation strategy. A full time-series stack of 22 Sentinel-
1A images was acquired during the 2016 growing season
(February-October) (Table 1). No Sentinel-1B imagery was
available over the study area. Level-1 ground-range detected,
descending mode, IW imagery acquired from the Alaska
Satellite Facility (a direct mirror of ESA’s Sci-hub) were
processed using the free and open source Sentinel-1 toolbox.
The ground-range detected images were processed following
guidelines including applying restituted orbit files, multi-look
azimuthal compressions to 20m, terrain correction using
SRTM 30m version 4 DEM, radiometric calibration
adjustments to correct for viewing geometry effects, and
refined lee speckle filter to reduce constructive and destructive
interference, all resulting in sigma-nought backscatter data
logarithmically scaled in dB [64]. Given the relatively small
study region with generally flat terrain, incident angle artifacts
and layover/shadow effects were relatively minimal. We
display incident angle and SAR backscatter ranges across the
study area in Table 1. We also highlight average SAR signal
and the temporal dynamic across the different land cover types
found in the study area (Fig. 1).
2) Training and Validation Data
We conducted fieldwork throughout the study area during
May/June and September/October of 2016 as part of a related
project on rice residue burning. The data included surveys on
crop calendar, crop rotation, field conditions, and biomass
data. Over 900 geolocated photos were taken of paddy fields
and non-rice areas such as aquaculture, wetlands, and other
land cover types found within the region. These field photos
were used for training or validation along with fine-resolution
imagery from Google Earth and the original Sentinel-1
imagery. No training data were included in the validation. We
show a multi-temporal composite of the SAR data with the
training polygons overlaid (Fig. 2).
III. METHODOLOGY
We classified the 6 different datasets (Table 2) using the
random forest algorithm to obtain rice areal estimates and as a
basis to compare the datasets. These resulting datasets were
the basis for comparison of the different polarizations and
spatial resolutions for mapping small-holder paddy rice.
Accordingly, we selected nominal 10m resolution as it’s the
native resolution of Sentinel-1 and 20m resolution data for
comparison. We addressed the following questions in our
study: 1) Which dataset yields the highest overall and class
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
specific accuracies and mapped areas, and are there notable
differences between them? 2) What is the typical rice
phenology including start/end of season and length of growing
season? 3) How do the resulting mapped rice area estimates
vary at the province and commune level?4) Based on
landscape analysis, what are the spatial characteristics of
paddy fields and approximately how many are in the study
area?
A. Double and single crop rice mapping
We employed an ensemble, machine-learning random forest
algorithm for mapping single and double crop paddy rice, as
well as for evaluating the relative importance of specific input
data for mapping [65]. The machine-learning random forest
algorithm implemented in Scikit-learn python package, uses
bootstrap aggregated sampling to build individual decision
trees for classification. Each decision tree was built with a
bootstrap sample from the training data, with the unsampled
data used for out-of-bag sampling. Within the structure of a
tree, a random sample of the square root of the number of
predictors was chosen for each split as best candidates derived
from the entire predictor set. Random forest is robust against
outliers and over-fitting, nonparametric, has high classification
accuracy, and can yield a measure of variable importance.
We implemented the random forest algorithm using all 22
(VV or VH) or 44 input bands (VV & VH) (Table 1) for each
of the six datasets separately. We populated the random forest
with 1000 trees, where out-of-bag errors reach asymptotic
values [65]. The same classification technique was applied to
each dataset. Out of bag samples, which are randomly
withheld from classification training were used as an indicator
of feature importance. From the 1000 trees populated in our
random forest, the mean decrease in classification accuracy for
the input bands was reported. This was useful as a measure of
feature importance in mapping paddy rice and each input band
can be linked to a general crop growth stage (i.e. sowing,
vegetative growth, and harvest). Training polygons were
digitized over the single rice, double rice, and non-rice areas
within each province using field photos, Sentinel-1 data,
survey information, and fine-resolution Google Earth imagery.
B. Accuracy assessment
We performed an accuracy assessment on the resulting rice
maps. Random points across the study area were generated
using a stratified random sampling scheme [66]. Based on the
proportion of each resulting mapped class, a stratified random
sample of 402 total polygons for double crop rice (125), single
crop rice (41) and non-rice (236) classes were generated for
the 10m data. For the 20m data the same polygons were used
for comparison consistency, however, there are fewer total
pixels due to reduced spatial resolution. We specified a 600m
minimum distance between polygons to prevent field overlap.
Following good practices in accuracy assessment we
adjusted the classification accuracies as well as the mapped
rice area estimates based on the weighting from the proportion
of land area for each class [67]. This weighting results in
unbiased areal estimates. Based on the same, we derived
uncertainty estimates for the accuracy-adjusted areas using a
95% confidence interval for each of our resulting mapped
classes. The accuracy-adjusted area provides a more robust
Copyright 2017 IEEE. Author submitted final version.
3
assessment of the area mapped for each class. We also
computed the confidence intervals for adjusted class and
overall accuracies to be used for comparison amongst the
different maps.
C. Spatial Analysis
We compared the paddy rice maps through a number of
metrics including Average Absolute Deviation (AAD) (1) and
Coefficient of Variation (CV) (2) to explore absolute and
relative variation in mapped area for each of the 6 different
datasets as follows:
∑|𝑥− 𝑥̅ |
𝐴𝐴𝐷 = (1)
𝑁
𝜎
𝐶𝑉 = ( ) 100 (2)
𝑥̅
Where x is the rice area in hectares for one of the six rice
maps, 𝑥̅ is the mean rice area of all 6 datasets, and N is the
number of datasets for (1). While for (2) σ is the standard
deviation of all rice area for all six datasets, and 𝑥̅ is the mean
rice area for all datasets. The final value is a percentage.
We compared the AAD and CV at the commune-level
(third-level administrative subdivision) and study-area level to
assess overall and spatial variability. In addition we computed
pixel-level thematic change for each dataset. We also
employed Fragstats (Version 4.2) for landscape-scale analysis
on the most accurate resulting dataset including number of
patches (collections of connected/adjacent paddy rice pixels)
and patch metrics useful for evaluating the size of large
connected aggregations/collections of paddy rice fields.
Further, we employed mean-shift image segmentation on the
double-rice pixels of the time-series imagery stack to estimate
rice collective size (groups of fields with similar crop
phenology) and number of fields in the study area by
including typical paddy field size of 790m2 [58]. We show the
general flowchart in Fig. 3. Mean-shift image segmentation is
a non-parametric iterative algorithm fitting a neighborhood
window around each pixel, calculating the data mean in the
window, and shifting the neighborhood window to the mean
[68], [69]. The algorithm is useful for clustering pixels with
similar signal and has been used in a variety of land cover
remote sensing applications [70], [71].
D. Rice phenology
While some of the prior studies utilized spatiotemporal
filtering to derive crop phenology metrics, we derived the
metrics based on the unique rice dynamic range combined
with general phenology timeframes. Sentinel-1 VH-polarized
backscatter imagery were used to estimate sowing
/transplanting and harvest dates for both seasons of rice. Based
on the unique phenology of paddy rice measured by SAR and
selecting a general timeframe based on a-priori knowledge, we
found local planting date coincides with the local minimum
value (indicative of flooding, constrained to February or
March), and harvest date when the local maximum backscatter
value is reached (indicative of peak maturity of rice just prior
to harvest, constrained to May or June) as noted in previous
studies [8], [41], [48], [72]. The estimated range is based on
the overpass dates of the Sentinel-1 satellite. The approximate
length of growing season is derived by differencing the
median date within the planting range and the harvest date
range.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
IV. RESULTS
A. Mapped area variation
The six paddy rice maps generated from the random forest
classifier and their spatial patterns are shown in Fig. 4. The
pixel classification count for double rice shown in Fig. 4
zoomed to a hotspot of variation area in Bac Ninh, highlights
that along edges and smaller fields some of the rice maps are
not in agreement. Overall, 66% of the pixels were in
agreement (where pixel count = 6). On another note, when the
absolute mapped rice areas are compared we notice more
variation. For example, the total mapped double-rice area for
each is: 208,276 ha (VV-10m), 212,465 ha (VH-10m),
214,565 ha (VV&VH-10m), 214,903 ha (VV-20m), 218,788
ha (VH-20m) and 220,356 ha (VV&VH-20m). This general
pattern suggests 10m imagery systematically reports lower
mapped area than 20m imagery, and VV reports less mapped
area than VH with VV&VH combination accounting for the
most area. This is attributed to VV signal attenuated by the
vertical structure of rice. The same pattern is not clear for
single rice. Whereas 10m datasets exhibited less area than
20m datasets, and reported areas varied between polarizations.
In comparison to the official government rice areal statistics
for 2016 of 232,700 ha paddy rice, our estimates reported
about 12,000 – 24,000 ha less [1]. In comparison to same
statistics for the entire Red River Delta (546,950 ha), our
study area has about 40% of the total rice area. The mapped
areas for single rice can be seen in Table 3.
B. Accuracy assessment and areal adjustment
We evaluated the accuracy of each dataset and present the
confusion matrices with the total number of pixels for each
class, as well as user, producer and overall accuracies adjusted
using the unbiased areal estimates. We also provide 95%
confidence intervals for each statistic for purposes of
comparison. The dataset overall accuracies in descending
order are VV&VH-10m (93.5%), VH-10m (93.1%), VV&VH-
20m (92.5%), VH-20m (91.4%), followed by VV-20m
(91.0%), and VV-10m (90.9%). We note the confidence
interval ranges between VV&VH-20m and VV&VH-10m, and
the same for the others, suggests they may not be significantly
different overall. However, when comparing class specific
user and producer accuracies for double crop rice and single
crop rice the differences are clear. The biggest difference is for
single crop rice where the 10m datasets have higher user and
producer accuracies (except VV). For double crop rice, the
user and producer accuracies are higher in the 10m than the
20m datasets. Across all datasets, user accuracies for single
and double crop rice are higher than producer accuracies
suggesting a net-positive of rice omission errors. Further
details on the confusion matrices are shown in Table 3.
We adjusted the total mapped areas based on the accuracy
assessment and resulting unbiased areal estimates for each
dataset and compared the trends as shown in Fig. 5. Overall,
we found that accuracy-adjusted areas for single and double
rice were higher than the mapped areas. This is due to the rice
omission errors found during the accuracy assessment. The
accuracy-adjusted areas yield an improved areal estimate
useful for studies requiring rice land area as a non-spatial
input. The 95% confidence intervals for each class are also
Copyright 2017 IEEE. Author submitted final version.
4
shown with Fig. 5 highlighting uncertainty in the mapped
areas.
C. Land cover and land use
We visually compared the 6 different datasets with fine-
resolution Google Earth imagery. We highlight one location
representative of the pattern throughout the study area (Fig. 6).
While all datasets capture the general rice spatial patterns
well, there are some errors and variation. Specifically, the
VH10m and VVVH-10m datasets are best able to reduce
commission errors from small irrigation ditches and roads
adjacent or within rice fields. All 20m datasets do not
effectively account for these areas regardless of polarization
due to spatial resolution. While it is technically error, we
highlight that this rice area commission error could be useful
for certain applications seeking to map not only rice land
cover, but also mapping rice land use. While these ditches and
roads are not rice, they are designed to support rice field
operations (threshing or drying of rice occurs in these areas)
and access to fields (such as for machine-harvester
equipment). However, while the 20m datasets are including
these adjacent roads or ditches in the rice category through
commission errors, they have fewer false small rice areas
(seen as small 5 or 6 pixel collections on the eastern sides of
the imagery). One reason for this is attributed to inherent noise
reduction through data processing involving multi-look
processing into 20m resolution. Thus, there are tradeoffs to
each resolution and polarization.
D. Pixel-level rice variation
Analysis of pixel-level thematic variation between each
dataset was conducted. For the 10m datasets we found the
biggest difference between VV and VH where 48,372 ha of
classified land area were different (6.29% difference) which is
much higher than total rice area variation (214,565 ha VH10m
vs 208,276ha VV10m). The main differences were attributed
to double-rice pixels converted to non-rice pixels, non-rice
converted to double rice, followed by non-rice into single rice.
The smallest difference was between VH and VV&VH with
12,343 ha (1.60% difference). In descending order, the
majority was attributed to non-rice to double rice, double rice
to non-rice, and non-rice to single rice. The 20m datasets
change was very similar with VV and VH (46,613 ha with
6.01% difference), VV and VV&VH (39,848 ha with 5.18%
difference), and VH and VV&VH (10,288 ha with 1.34%
difference). In all cases 20m datasets had about 0.2% less
difference than the 10m datasets, with each following a similar
pattern of variation.
E. Province-level rice variation
We evaluated the province-level variation in mapped rice
area between the different datasets through coefficient of
variation and z-scores for double and single rice (Fig. 7). The
CV results suggest relatively low and uniform variation in
double rice across the provinces with highest variation in Bac
Ninh (2.82%) and Hanoi (2.30%) followed by Ha Nam
(2.02%), Vinh Phuc (1.92%), and Hung Yen (1.77%). For
single rice the variation was wide ranging depending on the
province, with Vinh Phuc exhibiting the highest (11.57%)
variation followed by Bac Ninh (8.90%), Hanoi (8.37%), Ha
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
Nam (5.71%), and Hung Yen (4.17%). We attribute the
highest variation in Vinh Phuc to small-holder agriculture and
mosaicked landscape making classification difficult. The z-
score results for double-rice suggested more insight into
province-level variation. For VV20m Ha Nam and Hung Yen
had the highest positive z-scores whereas all other provinces
had negative Z-scores, and that VV had the most variation in
z-score across the provinces. This suggests the VV may be
more sensitive to different environmental parameters such as
water and moist soil underlying much of these areas. We note
similar Z-scores for VVVH20m with the least variation across
provinces in single and double crop rice. The same also goes
for VH20m with all z-scores for each dataset as positive.
Generally the 20m datasets tend to have more similar z-scores
than the 10m datasets.
F. Commune-level rice variation
For the entire study area the total mapped double rice area
for each dataset is relatively similar ranging from: 208,276ha
(VV10m) to 220,536ha (VVVH20m). However, results at the
commune level (third-level administrative subdivision)
suggest more variation. We note that while VV10m had the
overall lowest double-rice mapped area, the trend is not
spatially universal as VV10m had 27 communes with more
rice than VH10m. For example, VH10m in Loung Tai had
249ha less rice than VV10m with a percent difference of
4.60%. However, the biggest percent difference between the
two was in Ba Vi where VV10m had 20% less rice than
VH10m.
To better evaluate spatial variation at the commune level,
we measured the AAD and CV. Across all communes for each
dataset we found rice area did not vary uniformly across
space, based on the AAD and CV results (Fig. 8). We found
an average CV of 4.75% for each commune. The spatial
pattern suggested that much of the variation proportionately
occurred in the northern half of the study area (Vinh Phuc,
northern Hanoi, and northern Bac Ninh provinces), whereas
communes in Hung Yen and Ha Nam had variation between
the datasets almost always less than 2.5%. We also note that
the communes surrounding Hanoi City had relatively higher
CV, but low AAD due to small amount of rice in these areas.
We attribute the spatial clustering of variation to land cover
and land use differences, and the difficulty of mapping in a
small-holder mosaicked landscape. Accordingly, the northern
areas dominated by more wetlands, hills, and especially
smaller rice fields intermixed with small-holder non-rice crops
are likely to have greater class confusion.
G. Rice phenology
We estimated the length of growing season, start of season,
and end of season for rice based on the SAR time-series and
the unique dynamic range in SAR signal found for rice
paddies (Fig. 9). The results were aggregated to commune
level for comparison. Across all communes we found an
average length of the first rice season to be 93 days, while the
second season was approximately 104 days. We attribute the
relatively lower range in season 1 due to most fields remaining
flooded for several months from the previous season harvest
and missing SAR data for ~June 28th. In reality, the first
season length is likely slightly underestimated. In addition
Copyright 2017 IEEE. Author submitted final version.
5
these are approximations dependent on the Sentinel-1 overpass
dates (+-11 days). The estimated average sowing date for the
2nd season is Day of Year (DoY) 175 (23rd of June) and DoY
55 (24th of February) for the first season. For the harvest we
found an average DoY of 148 (27th of May) for season 1.
Whereas for season 2 we found estimated average harvest of
DoY 279 (5th of October). These estimates are generally in
line with our local knowledge and field experience. A more
precise estimate of length of season, harvest, and sowing could
be obtained by including both Sentinel-1A and Sentinel-1B,
cutting the overpass time in half.
H. Rice landscape metrics
Fragstats analysis using the most accurate VVVH10m map
produced a total of 31,934 large continuous tracts/patches of
double-rice pixels with an average area of 6.72ha (σ=59.4ha)
and large standard deviation. Mean-shift image segmentation
combined with fragstats analysis on the time-series stack
masked for double-rice pixels, suggests 796,720 rice
collectives (small patches of rice with similar SAR signal)
with an average area of 2668m2 (σ=7250m2). The average
field size is reported to be 790m2 (σ =625m2) [58], thus an
average of 3.4 individual rice fields make up a typical rice
collective which follow a similar rice phenology pattern
including harvest and flooding dates, due to similarity in
image segmentation. Based on the aforementioned average
field size and total double-rice area from the VVVH10m
scenario we estimate 2.69 million double-rice paddy fields
within the HCR.
I. Input band importance
The feature importance of each input band for the paddy
rice classification was assessed in the random forest classifier
through mean decrease in accuracy (Fig. 10). We found that
the top 3 most important bands for VH(10m and 20m), and
VV(20m) were generally the same with bands 11, 12, and 13
(Fig. 10). These three bands were from June and early July
(Table 1) suggesting that imagery from spring harvest and
summer planting season are most important as removing one
of these bands results in a mean decrease in accuracy of 3-6%
depending on the dataset. In addition, these bands are
important for discriminating between single and double crop
rice. Other bands obtained during planting, growth, and
harvest, however, are also important as removing any one of
them still reduces overall accuracy significantly. For VV (10m
only) we found bands 7, 6, and 8 to be most important. These
bands are from the main vegetative growth stage for winter
rice (April). A relatively similar trend was observed in VVVH
datasets. The average local incidence angle and standard
deviation over rice areas suggests minimal variation in
incidence angle (µ = 38.63degrees, σ = 3.34degrees). Thus, it
is not critical to account for the incidence angle variation
effects on backscatter in the study area (Table 1) due to
minimal variation. In addition, the time-series signal of double
rice versus single rice shows it is possible to differentiate
between the two with only the first season of data. Double rice
areas have relatively higher backscatter throughout the first
rice season (Fig. 1).
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
V. DISCUSSION
Early rice mapping studies in Asia employing SAR have
had varying degrees of success and accuracy. Some of the
earliest studies using RADARSAT or ERS-1 found good
agreement between their rice maps and existing validation
data, with the latter study reporting overall accuracy of 87%
for a small study area and coarse resolution imagery [5], [8],
[11]. More recent studies employing C-band RADARSAT
imagery have reported rice class accuracies as high as 98%
and 90% using TerraSAR-X imagery [30], [72]. Studies have
also highlighted advantage of multi-date versus single-date
imagery for rice mapping [28]. However, in Vietnam the case
is different. In the Mekong River Delta using multi-temporal
SAR including ENVISAT and Sentinel-1, studies have found
overall accuracy for land cover or rice maps ranging from
80% - 92% [20], [21], [23], [31], [33], [35] [73]. In the Red
River Delta, using multitemporal SAR such as RADARSAT-2
and ENVISAT, studies have found overall accuracy for rice
maps ranging from 71-95% [29], [41], [74], [75]; and overall
accuracy of 89% in Vietnam with TerraSAR-X [27].
Our accuracy assessment found overall strong user’s
accuracies for most datasets and classes with double rice
ranging from 92.1% - 96.9% and single rice (95.2% - 100%).
Whereas, producer’s accuracies were generally lower (77.8% -
83.5% double rice, and 35.0% - 59.0% single rice) with
overall accuracies between 90.9% - 93.5%.
Our study findings are important in that the different spatial
resolution or polarization combinations have significant
impacts on mapped rice area. While these differences may not
be high at the study area scale, the difference between the
datasets is high at the commune and pixel level with CV
averaging about 10% at the commune scale. The most
variation was found in communes in the northern half of the
study area in Northern Hanoi, Vinh Phuc, and Bac Ninh where
small-holder farms in mosaicked landscapes are dominant,
making mapping difficult [2], [74]. These spatial differences
are important for different rice mapping applications where
spatial location is critical, such as rice straw burning emissions
and air quality assessments. We highlight the importance of
selecting the most accurate map which in this case is
VVVH10m. We also demonstrate though comparison at
different spatial scales that 10m data has more variation in rice
area compared to 20m data. We find that 20m data may prove
effective for land use mapping due to inclusion of a variety of
classes such as irrigation canals, and adjacent dirt farm roads,
whereas 10m may be best for specific thematic (rice in our
case) mapping. Our study also found that VH polarization
were more accurate than VV polarizations for rice mapping.
This has been attributed to VV more influenced by standing
water in fields and the signal attenuated by vertical structure
of rice [21], [35]. Whereas VH is less affected and more
representative of the actual rice growth and plant canopy
structure [33]. We also found our map had slightly more rice
area than the official government data, and other rice mapping
studies have found good agreement between maps and
government data in Vietnam [60], [74].
Using a simple method to replicate metrics on rice
phenology, we demonstrate the utility of our phenology
estimates for regional and global rice monitoring systems such
Copyright 2017 IEEE. Author submitted final version.
6
as Asia-RiCE or GEOGLAM. Additional applications could
improve upon existing SAR and MODIS disaster management
strategies with improved phenology metrics [76]. This is also
demonstrated through random forest feature importance
metrics that reducing some input data (i.e. near harvest) can
yield slightly less accurate maps as they may not capture the
wide backscatter range useful to distinguish paddy rice from
other land cover classes. However, these maps are still useful
for operational monitoring applications.
Combining our most accurate rice map with image
segmentation, Fragstats landscape analysis and field data, we
estimated the total number of paddy rice fields in the study
area (2.69million). In addition we found the average size of
continuous rice area patches, as well as the estimate of the
average size of a rice collective (2668m2, or 3.4 rice fields).
As mentioned, the rice collectives are continuous rice areas
undergoing similar rice phenology (i.e. managed in-sync).
While other classification methods could have been used we
chose random forest as it is robust, widely applied, generally
has similar results in agricultural applications [43], is less of a
black box, and computes faster than other algorithms such as
support vector machines or artificial neural networks [77].
Ensemble-mean classifiers such as random forest also
outperform standard approaches such as single decision tree or
maximum-likelihood classifier [42]. Random forest classifier
is also used for reducing feature inputs to reduce data
dimensionality, processing time, and redundancy [78], [79],
whereas for this study we maintained all inputs as we already
needed the full time-series of SAR for growing season
estimates.
Our study was limited due to several factors. The
availability of Sentinel-1A, but not 1B, limits data input to
approximately 12 day time intervals. At this temporal
frequency, it can be difficult to fully capture distinct crop
periods of initial inundation or harvest due to very active field
management. Increased availability of Sentinel-1B, however,
will enhance phenology monitoring. The rice phenology
estimates are limited by the same satellite overpass dates. In
general our length of season estimates for season 1 seemed
slightly lower than our local knowledge of about 100-115 days
due to missing SAR data on June 28th (as some areas harvest
around this time). However, it still highlights the utility of
SAR for monitoring the length of season without the need to
employ a complex algorithm or filtering.
We attribute relatively low single-rice producer’s accuracy
to the limited availability of training and validation data. This
land cover class occupies less than 1% of the landscape and
we could only obtain 41 polygons for validation. As such,
when we adjusted the producer’s accuracy to get unbiased
areal estimates, the accuracy decreased significantly. For
example, VH10m un-adjusted producer’s accuracy for single
rice is 97.0% with only 5 pixels out of 166 confused with
other classes. However, after calculating the accuracy using
unbiased areal estimates, the accuracy decreases to 59%.
Whereas for double rice, the changes are less dramatic (i.e.
just a few percentage points) due to appropriate sample size.
With further analysis, the results could be useful for land
cover versus land use mapping. In addition, based on the SAR
signal, it is possible to map single rice and double rice areas
with only the first season of data. Results could be useful for
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
aquaculture applications and land use planning. It may also be
beneficial to integrate moderate resolution Sentinel-1 SAR
imagery with high resolution UAV-obtained imagery for
relating to monitoring and improving crop growth parameter
estimation [80].
VI. CONCLUSION
Using a random forest classifier, we compared single and
double crop paddy rice mapping using 6 different datasets
based on varied polarizations (VV, VH, and both) as well as
different spatial resolutions (10m and 20m). Local incidence
angles for the complete SAR time-series dataset suggest that
imagery can be obtained for the whole year with negligible
variation reducing the need to account for angular influence on
backscatter. The time-series Sentinel-1 SAR imagery was able
to estimate start of season, end of season, and length of season
for each rice crop at the commune level with promising
results. Rice mapping estimates for the 6 datasets suggested
the most accurate dataset to be VVVH10m (93.5% OA),
whereas 20m datasets generally had lower overall, user’s and
producer’s accuracies compared to the 10m datasets. Thematic
change between the datasets suggested the least spatial
variation in classes to be between VH and VVVH (10m and
20m). The most common difference was between double crop
rice and non-rice areas. 20m datasets overestimated rice land
area more compared with 10m datasets, as pixels were too
large to capture small roads and ditches within paddy fields.
The 20m datasets, however, could be well-representative of
not only rice land cover, but especially rice land use.
Province-level variation in mapped rice areas suggested the
most occurring with VV20m, and also more variation in single
rice than double rice based on the coefficient of variation.
Moderately-high spatial variation was found at the commune-
level with the maximum variation in rice area as 20% with
VV10m and VH10m. This spatial variation could be critical
for spatially-explicit land cover mapping applications.
Average absolute deviation and coefficient of variation results
suggest communes in the northern portion of the study area
and those immediately surrounding Hanoi City have the most
proportional variation. This is important as these areas are
subject to peri-urban expansion and important for air quality
studies as areas closest to the city could have the highest
impact on urban air quality.
Overall our study compared six different datasets (VV10m,
VV20m, VH10m, VH20m, VVVH10m, and VVVH20m) for
mapping paddy rice, evaluated importance of input bands from
different dates for classification, estimated rice phenology and
calculated rice landscape metrics. The results from our study
provide useful insights for crop mapping/monitoring efforts
using SAR datasets in general and particularly within the Red
River Delta of Vietnam.
VII. ACKNOWLEDGEMENT
This work was supported by the University of Maryland,
Council on the Environment Green Fund Fellowship. K.
Lasko was partially supported by the Science, Mathematics
and Research for Transformation (SMART) scholarship by the
American Society for Engineering Education. The authors
thank Thanh Nhat Thi Nguyen and Hung Quang Bui (Vietnam
Copyright 2017 IEEE. Author submitted final version.
7
National University Hanoi) for field assistance, and William
Salas and Nathan Torbick (Applied Geosolutions, LLC) for
SAR expertise. K. Lasko thanks Michael Campbell
(Geospatial Research Lab) for support and also thanks his PhD
committee members including Christopher Justice, Krishna
Vadrevu, Ivan Csiszar, Louis Giglio, and Matthew Hansen for
support. Lastly, we thank the European Space Agency for the
open data policy and Alaska Satellite Facility for hosting a
mirror of the Sentinel-1 data.
VIII. REFERENCES
[1] General Statistics Office of Vietnam, 2016. Agriculture, Forestry, and
Trade. Accessed from:
http://www.gso.gov.vn/Default_en.aspx?tabid=491.
[2] Pham, V.C., Pham, T.T.H., Tong, T.H.A., Nguyen, T.T.H. and Pham,
N.H., 2015. The conversion of agricultural land in the peri-urban areas
of Hanoi (Vietnam): patterns in space and time. J. Land Use Sci., 10(2),
pp.224-242.
[3] Whitcraft, A.K., Vermote, E.F., Becker-Reshef, I. and Justice, C.O.,
2015. Cloud cover throughout the agricultural growing season: Impacts
on passive optical earth observations. Rem. Sens. Enviro., 156, 438-447.
[4] Paudyal, D.R. and Aschbacher, J., 1993. Land-cover separability studies
of filtered ERS-1 SAR images in the tropics. In Geoscience and Remote
Sensing Symposium, IGARSS'93. Better Understanding of Earth
Environment., International 1216-1218.
[5] Aschbacher, J., Pongsrihadulchai, A., Karnchanasutham, S., Rodprom,
C., Paudyal, D.R. and Le Toan, T., 1995. Assessment of ERS-1 SAR
data for rice crop mapping and monitoring. In Geoscience and Remote
Sensing Symposium, 1995. IGARSS'95.'Quantitative Remote Sensing for
Science and Applications', International Vol. 3, 2183-2185.
[6] Takeuchi, S. and Suwanwerakamtorn, R., 1996. Monitoring of Land
Cover Conditions in Paddy Fields Using Multitemporal SAR Data. Int.
Archives of Photogramm. And Rem. Sens., 31, 683-688.
[7] Chakraborty, M., Panigrahy, E.S. and Sharma, S.A., 1997.
Discrimination of rice crop grown under different cultural practices
using temporal ERS-1 synthetic aperture radar data. ISPRS J.
Photogramm. Remote Sens., 52(4), 183-191.
[8] Le Toan, T., Ribbes, F., Wang, L.F., Floury, N., Ding, K.H., Kong, J.A.,
Fujita, M. and Kurosu, T., 1997. Rice crop mapping and monitoring
using ERS-1 data based on experiment and modeling results. IEEE
Transactions on Geoscience and Remote Sensing, 35(1), 41-56.
[9] Liew, S.C., Kam, S.P., Tuong, T.P., Chen, P., Minh, V.Q. and Lim, H.,
1998. Application of multitemporal ERS-2 synthetic aperture radar in
delineating rice cropping systems in the Mekong River Delta,
Vietnam. IEEE Trans. Geosci. Rem. Sens., 36(5), 1412-1420.
[10] Yun, S., Cuizhen, W., Xiangtao, F. and Hao, L., 1997, August.
Estimation of rice growth status using RADARSAT data. In Geoscience
and Remote Sensing, 1997. IGARSS'97. Remote Sensing-A Scientific
Vision for Sustainable Development., 1997 IEEE International (Vol. 3,
1430-1432). IEEE.
[11] Ribbes, F., 1999. Rice field mapping and monitoring with RADARSAT
data. Int. J. Rem. Sens., 20(4), 745-765.
[12] Panigrahy, S., Manjunath, K.R., Chakraborty, M., Kundu, N. and
Parihar, J.S., 1999. Evaluation of RADARSAT standard beam data for
identification of potato and rice crops in India. ISPRS J. Photogramm.
Remote Sens., 54(4), 254-262.
[13] Shao, Y., Fan, X., Liu, H., Xiao, J., Ross, S., Brisco, B., Brown, R. and
Staples, G., 2001. Rice monitoring and production estimation using
multitemporal RADARSAT. Rem. Sens. Enviro., 76(3), 310-325.
[14] Lee, K. S., and S. I. Lee. "Assessment of post-flooding conditions of rice
fields with multi-temporal satellite SAR data." Int. j. Rem. sens. 24, no.
17 (2003): 3457-3465.
[15] Suga, Y., Takeuchi, S., Oguro, Y. and Konishi, T., 2000. Monitoring of
rice-planted areas using space-borne SAR data. Int. Arch. of
PHOTOGRAM. and Rem. Sens., 33(B7/4; PART 7), pp.1480-1483.
[16] Inoue, Y., Kurosu, T., Maeno, H., Uratsuka, S., Kozu, T., Dabrowska-
Zielinska, K. and Qi, J., 2002. Season-long daily measurements of
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
multifrequency (Ka, Ku, X, C, and L) and full-polarization backscatter
signatures over paddy rice field and their relationship with biological
variables. Rem. Sens. of Enviro., 81(2), 194-204.
[17] Chen, C. and McNairn, H., 2006. A neural network integrated approach
for rice crop monitoring. Int. J. Rem. Sens., 27(7), pp.1367-1393.
[18] Salas, W., Boles, S., Li, C., Yeluripati, J.B., Xiao, X., Frolking, S. and
Green, P., 2007. Mapping and modelling of greenhouse gas emissions
from rice paddies with satellite radar observations and the DNDC
biogeochemical model. Aquatic conservation: Marine and freshwater
ecosystems, 17(3), 319-329.
[19] Zhang, Y., Wang, C., Wu, J., Qi, J. and Salas, W.A., 2009. Mapping
paddy rice with multitemporal ALOS/PALSAR imagery in southeast
China. Int. J. Rem. sens., 30(23), 6301-6315.
[20] Lam-Dao, N., Apan, A., Young, F.R., Le-Van, T., Le-Toan, T. and
Bouvet, A., 2007, November. Rice monitoring using ENVISAT-ASAR
data: preliminary results of a case study in the Mekong River Delta,
Vietnam. In Proceedings of the 28th Asian Conference on Remote
Sensing (ACRS 2007). Geoinformatics Center (Klong Luang).
[21] Bouvet, A., Le Toan, T. and Lam-Dao, N., 2009. Monitoring of the rice
cropping system in the Mekong Delta using ENVISAT/ASAR dual
polarization data. IEEE Trans. on Geosci. Rem. Sens., 47(2), 517-526.
[22] Jia, K., Li, Q., Tian, Y., Wu, B., Zhang, F. and Meng, J., 2012. Crop
classification using multi-configuration SAR data in the North China
Plain. Int.l J. Rem. Sens., 33(1), pp.170-183.
[23] Nguyen, D.B., Clauss, K., Cao, S., Naeimi, V., Kuenzer, C. and Wagner,
W., 2015. Mapping rice seasonality in the Mekong Delta with multi-year
Envisat ASAR WSM data. Rem. Sens., 7(12), pp.15868-15893.
[24] Pei, Z., Zhang, S., Guo, L., McNairn, H., Shang, J. and Jiao, X., 2011.
Rice identification and change detection using TerraSAR-X
data. Canadian J. Rem. Sens., 37(1), 151-156.
[25] Gebhardt, S., Huth, J., Nguyen, L.D., Roth, A. and Kuenzer, C., 2012. A
comparison of TerraSAR-X Quadpol backscattering with RapidEye
multispectral vegetation indices over rice fields in the Mekong Delta,
Vietnam. Int. j. rem. sens., 33(24), pp.7644-7661.
[26] Inoue, Y., Sakaiya, E. and Wang, C., 2014. Capability of C-band
backscattering coefficients from high-resolution satellite SAR sensors to
assess biophysical variables in paddy rice. Rem. Sens. Enviro., 140,
pp.257-266.
[27] Nelson, A., Setiyono, T., Rala, A.B., Quicho, E.D., Raviz, J.V.,
Abonete, P.J., Maunahan, A.A., Garcia, C.A., Bhatti, H.Z.M., Villano,
L.S. and Thongbai, P., 2014. Towards an operational SAR-based rice
monitoring system in Asia: Examples from 13 demonstration sites across
Asia in the RIICE project. Rem. Sens., 6(11), pp.10773-10812.
[28] Jiao, X., Kovacs, J.M., Shang, J., McNairn, H., Walters, D., Ma, B. and
Geng, X., 2014. Object-oriented crop mapping and monitoring using
multi-temporal polarimetric RADARSAT-2 data. ISPRS J. Photogram.
Rem. Sens., 96, pp.38-46.
[29] Hoang, H.K., Bernier, M., Duchesne, S. and Tran, Y.M., 2016. Rice
Mapping Using RADARSAT-2 Dual-and Quad-Pol Data in a Complex
Land-Use Watershed: Cau River Basin (Vietnam). IEEE J-STARS.
g, 9(7), 3082-3096.
[30] Asilo, S., de Bie, K., Skidmore, A., Nelson, A., Barbieri, M. and
Maunahan, A., 2014. Complementarity of two rice mapping approaches:
Characterizing strata mapped by hypertemporal MODIS and rice paddy
identification using multitemporal SAR. Rem. Sens., 6(12), 12789-
12814.
[31] Karila, K., Nevalainen, O., Krooks, A., Karjalainen, M. and
Kaasalainen, S., 2014. Monitoring changes in rice cultivated area from
SAR and optical satellite images in Ben Tre and Tra Vinh Provinces in
Mekong Delta, Vietnam. Rem. Sens., 6(5), pp.4090-4108.
[32] Torbick, N., Chowdhury, D., Salas, W. and Qi, J., 2017. Monitoring
Rice Agriculture across Myanmar Using Time Series Sentinel-1
Assisted by Landsat-8 and PALSAR-2. Rem. Sens., 9(2), p.119.
[33] Nguyen, D.B., Gruber, A. and Wagner, W., 2016. Mapping rice extent
and cropping scheme in the Mekong Delta using Sentinel-1A data. Rem.
Sens. Lett., 7(12), pp.1209-1218.
[34] Mansaray, L.R., Zhang, D., Zhou, Z. and Huang, J., 2017. Evaluating
the potential of temporal Sentinel-1A data for paddy rice discrimination
at local scales. Rem. Sens. Lett., 8(10), pp.967-976.
Copyright 2017 IEEE. Author submitted final version.
8
[35] Son, N.T., Chen, C.F., Chen, C.R. and Minh, V.Q., 2017. Assessment of
Sentinel-1A data for rice crop classification using random forests and
support vector machines. Geocarto Int., pp.1-15.
[36] Steele-Dunne, S.C., McNairn, H., Monsivais-Huertero, A., Judge, J.,
Liu, P.W. and Papathanassiou, K., 2017. Radar remote sensing of
agricultural canopies: A review. IEEE J-STARS, 10(5), pp.2249-2273.
[37] Whitcomb, J., Moghaddam, M., McDonald, K., Kellndorfer, J. and
Podest, E., 2009. Mapping vegetated wetlands of Alaska using L-band
radar satellite imagery. Canadian J. of Rem. Sens., 35(1), 54-72.
[38] White, L., Brisco, B., Dabboor, M., Schmitt, A. and Pratt, A., 2015. A
collection of SAR methodologies for monitoring wetlands. Rem.
sens., 7(6), pp.7615-7645.
[39] Wang, W., Yang, X., Li, X., Chen, K., Liu, G., Li, Z. and Gade, M.,
2017. A Fully Polarimetric SAR Imagery Classification Scheme for
Mud and Sand Flats in Intertidal Zones. IEEE Trans. Geosci. Rem.
Sens., 55(3), pp.1734-1742.
[40] Shiraishi, T., Motohka, T., Thapa, R.B., Watanabe, M. and Shimada, M.,
2014. Comparative assessment of supervised classifiers for land use–
land cover classification in a tropical region using time-series palsar
mosaic data. IEEE J-STARS, 7(4), pp.1186-1199.
[41] Torbick, N., Salas, W., Chowdhury, D., Ingraham, P. and Trinh, M.,
2017. Mapping rice greenhouse gas emissions in the Red River Delta,
Vietnam. Carbon Management, 8(1), pp.99-108.
[42] Waske, B. and Braun, M., 2009. Classifier ensembles for land cover
mapping using multitemporal SAR imagery. ISPRS Journal of
Photogrammetry and Remote Sensing, 64(5), pp.450-457.
[43] Sonobe, R., Tani, H., Wang, X., Kobayashi, N. and Shimamura, H.,
2014. Random forest classification of crop type using multi-temporal
TerraSAR-X dual-polarimetric data. Rem. Sens. Lett., 5(2), 157-164.
[44] Zhang, H., Li, Q., Liu, J., Du, X., Dong, T., McNairn, H., Champagne,
C., Shang, J. and Liu, M., 2017. Object-based crop classification using
multi-temporal SPOT-5 imagery and textural features with a Random
Forest classifier. Geocarto International, pp.1-40.
[45] Uhlmann, S. and Kiranyaz, S., 2014. Integrating color features in
polarimetric SAR image classification. IEEE Transactions on
Geoscience and Remote Sensing, 52(4), pp.2197-2216.
[46] Lopez-Sanchez, J.M., Cloude, S.R. and Ballester-Berman, J.D., 2012.
Rice phenology monitoring by means of SAR polarimetry at X-
band. IEEE Trans. Geosci. Rem. Sens., 50(7), 2695-2709.
[47] Jiao, X., Kovacs, J.M., Shang, J., McNairn, H., Walters, D., Ma, B. and
Geng, X., 2014. Object-oriented crop mapping and monitoring using
multi-temporal polarimetric RADARSAT-2 data. ISPRS J. Photogram.
Rem. Sens., 96, pp.38-46.
[48] Koppe, W., Gnyp, M.L., Hütt, C., Yao, Y., Miao, Y., Chen, X. and
Bareth, G., 2013. Rice monitoring with multi-temporal and dual-
polarimetric TerraSAR-X data. Int. J. Appl. Earth Obs. Geoinfo., 21,
pp.568-576.
[49] Erten, E., Rossi, C., Yüzügüllü, O. and Hajnsek, I., 2014, July.
Phenological growth stages of paddy rice according to the BBCH scale
and SAR images. In Geosci. and Rem. Sens. Symp. (IGARSS), 2014
IEEE International (pp. 1017-1020). IEEE.
[50] Erten, E., Rossi, C. and Yüzügüllü, O., 2015. Polarization impact in
TanDEM-X data over vertical-oriented vegetation: The paddy-rice case
study. IEEE Geosci. Rem. Sens. Lett., 12(7), 1501-1505.
[51] Lopez-Sanchez, J.M., Vicente-Guijalba, F., Ballester-Berman, J.D. and
Cloude, S.R., 2014. Polarimetric response of rice fields at C-band:
Analysis and phenology retrieval. IEEE Trans. Geosci. Rem.
Sens., 52(5), pp.2977-2993.
[52] Küçük, Ç., Taşkın, G. and Erten, E., 2016. Paddy-rice phenology
classification based on machine-learning methods using multitemporal
co-polar X-band SAR images. IEEE J-STARS, 9(6), pp.2509-2519.
[53] de Bernardis, C.G., Vicente-Guijalba, F., Martinez-Marin, T. and Lopez-
Sanchez, J.M., 2015. Estimation of key dates and stages in rice crops
using dual-polarization SAR time series and a particle filtering
approach. IEEE J-STARS, 8(3), pp.1008-1018.
[54] Xie, L., Zhang, H., Wu, F., Wang, C. and Zhang, B., 2015. Capability of
rice mapping using hybrid polarimetric SAR data. IEEEJ-STARS, 8(8),
pp.3812-3822.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
[55] Yuzugullu, O., Erten, E. and Hajnsek, I., 2015. Rice growth monitoring
by means of X-band co-polar SAR: Feature clustering and BBCH
scale. IEEE Geosci. Rem. Sens. Lett., 12(6), 1218-1222.
[56] Oyoshi, K., Tomiyama, N., Okumura, T., Sobue, S. and Sato, J., 2016.
Mapping rice-planted areas using time-series synthetic aperture radar
data for the Asia-RiCE activity. Paddy and Water Environment, 14(4),
pp.463-472.
[57] Whitcraft, A.K., Becker-Reshef, I. and Justice, C.O., 2015. A framework
for defining spatially explicit earth observation requirements for a global
agricultural monitoring initiative (GEOGLAM). Remote Sensing, 7(2),
pp.1461-1481.
[58] Lasko, K., Vadrevu, K., Tran, V., Ellicott, E., Nguyen, T., Bui, H. and
Justice, C., 2017. Satellites may underestimate rice residue and
associated burning emissions in Vietnam. Enviro. Res. Lett., 12(8),
085006.
[59] Xiao, X., Boles, S., Liu, J., Zhuang, D., Frolking, S., Li, C., Salas, W.
and Moore, B., 2005. Mapping paddy rice agriculture in southern China
using multi-temporal MODIS images. Rem. Sens. Enviro., 95(4),
pp.480-492.
[60] Kontgis, C., Schneider, A. and Ozdogan, M., 2015. Mapping rice paddy
extent and intensification in the Vietnamese Mekong River Delta with
dense time stacks of Landsat data. Rem. Sens. Enviro., 169, 255-269.
[61] Duong, P.T. and Yoshiro, H., 2015. Current Situation and Possibilities
of Rice Straw Management in Vietnam. Accessed from:
http://www.jsrsai.jp/Annual_Meeting/PROG_52/ResumeC/C02-4.pdf.
[62] Dey M.M. and Ahmed, M., 2005. Aquaculture—food and livelihoods
for the poor in Asia: a brief overview of the issues. Aquaculture
Economics & Management, 9(1-2), pp.1-10.
[63] Ottinger, M., Clauss, K. and Kuenzer, C., 2017. Large-Scale Assessment
of Coastal Aquaculture Ponds with Sentinel-1 Time Series Data. Rem.
Sens., 9(5), p.440.
[64] Zuhlke, M., Fomferra, N., Brockmann, C., Peters, M., Veci, L., Malik, J.
and Regner, P., 2015, December. SNAP (Sentinel Application Platform)
and the ESA Sentinel 3 Toolbox. In Sentinel-3 for Science
Workshop (Vol. 734, p. 21).
[65] Breiman, L., 2001. Random forests. Machine learning, 45(1), pp.5-32.
[66] Congalton, R.G. and Green, K., 2008. Assessing the accuracy of
remotely sensed data: principles and practices. CRC press.
[67] Olofsson, P., Foody, G.M., Herold, M., Stehman, S.V., Woodcock, C.E.
and Wulder, M.A., 2014. Good practices for estimating area and
assessing accuracy of land change. Rem. Sens. Enviro., 148, 42-57
[68] Fukunaga, K. and Hostetler, L., 1975. The estimation of the gradient of a
density function, with applications in pattern recognition. IEEE
Transactions on information theory, 21(1), pp.32-40.
[69] Comaniciu, D. and Meer, P., 2002. Mean shift: A robust approach
toward feature space analysis. IEEE Transactions on pattern analysis
and machine intelligence, 24(5), pp.603-619.
[70] Su, T. and Zhang, S., 2017. Local and global evaluation for remote
sensing image segmentation. ISPRS Journal of Photogrammetry and
Remote Sensing, 130, pp.256-276.
[71] Senthilnath, J., Omkar, S.N., Mani, V., Tejovanth, N., Diwakar, P.G.
and Shenoy, A.B., 2012. Hierarchical clustering algorithm for land cover
mapping using satellite images. IEEE J-STARS, 5(3), pp.762-768.
[72] Choudhury, I. and Chakraborty, M., 2006. SAR signature investigation
of rice crop using RADARSAT data. Int. J. Rem. Sens., 27(3), pp.519-
534.
[73] Chen, C.F., Son, N.T., Chen, C.R., Chang, L.Y. and Chiang, S.H., 2016.
Rice crop mapping using Sentinel-1A phonological metrics. ISPRS-
International Archives of the Photogrammetry,
[74] Nguyen, D., Wagner, W., Naeimi, V. and Cao, S., 2015. Rice-planted
area extraction by time series analysis of ENVISAT ASAR WS data
using a phenology-based classification approach: A case study for Red
River Delta, Vietnam. The International Archives of Photogrammetry,
Remote Sensing and Spatial Information Sciences, 40(7), p.77.
[75] Hoang, K.H., Bernier, M., Duchesne, S. and Tran, M.Y., 2014, July.
Classification of rice fields in a complex land-use watershed in Northern
Vietnam using RADARSAT-2 data. In Geoscience and Remote Sensing
Symposium (IGARSS), 2014 IEEE International (pp. 1501-1503). IEEE.
Copyright 2017 IEEE. Author submitted final version.
9
[76] Boschetti, M., Nelson, A., Nutini, F., Manfron, G., Busetto, L., Barbieri,
M., Laborte, A., Raviz, J., Holecz, F., Mabalay, M.R.O. and Bacong,
A.P., 2015. Rapid assessment of crop status: An application of MODIS
and SAR data to rice areas in Leyte, Philippines affected by Typhoon
Haiyan. Remote Sensing, 7(6), pp.6535-6557.
[77] Belgiu, M. and Drăguţ, L., 2016. Random forest in remote sensing: A
review of applications and future directions. ISPRS J. Photogramm.
Remote Sens., 114, pp.24-31.
[78] Loosvelt, L., Peters, J., Skriver, H., De Baets, B. and Verhoest, N.E.,
2012. Impact of reducing polarimetric SAR input on the uncertainty of
crop classifications based on the random forests algorithm. IEEE
Transactions on Geoscience and Remote Sensing, 50(10), pp.4185-4200.
[79] Rodriguez-Galiano, V.F., Chica-Olmo, M., Abarca-Hernandez, F.,
Atkinson, P.M. and Jeganathan, C., 2012. Random Forest classification
of Mediterranean land cover using multi-seasonal imagery and multi-
seasonal texture. Rem. Sens. Environ., 121, pp.93-107.
[80] Uto, K., Seki, H., Saito, G. and Kosugi, Y., 2013. Characterization of
rice paddies by a UAV-mounted miniature hyperspectral sensor
system. IEEE J-STARS, 6(2), pp.851-860.
Kristofer Lasko received the B.S. degree in
geographical sciences and M.S. degree in
geospatial information science from the
University of Maryland, College Park, MD,
USA in 2012 and 2014 respectively. He is
currently pursuing the PhD degree focusing
on estimating rice residue burning and
emissions variation through integrating field
and remote sensing data with a focus in the Vietnam region.
He has worked as a research intern and project lead at two
different NASA locations. He subsequently worked for the
University of Maryland as a Geospatial Scientist while
pursuing his master’s degree. He is currently a graduate
research assistant at the University of Maryland and on a
scholarship from the American Society for Engineering
Education.
Krishna P. Vadrevu received the PhD
degree in ecology and remote sensing in
2000 while working at the Indian Space
Research Organization (ISRO), National
Remote Sensing Center, Hyderabad, India.
His thesis focused on biodiversity and forest
characterization using remote sensing
techniques.
From 2001 to 2003, he worked as a Post-doctoral
Researcher and 2004-2009 as a Research Scientist at the Ohio
State University, USA. From 2010-2016 he worked as an
Associated Research Professor with the Department of
Geographical Sciences, University of Maryland college park,
USA. He joined the NASA Marshall Space Flight Center in
Huntsville, Alabama, USA as a remote sensing scientist
during July, 2016. He also serves as the lead scientist for
NASA South/Southeast Research Initiative
(www.sari.umd.edu) and is on the editorial board for five
different peer-reviewed journals.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 10

Vinh T Tuan received the B.S. degree in

mathematics – informatics education from
Hanoi Pedagogical University 2 in Vinh Phuc,
and the M.S. degree in information technology
from Hanoi University of Science and
Technology in Hanoi, Vietnam. He is
currently pursuing the PhD degree in remote
sensing at Vietnam National University: Hanoi.
Since B.S. graduation in 1998 he has served as a lecturer at
the Hanoi Pedagogical University 2. He is also a part-time
PhD student researcher at the University of Engineering and
Technology, Vietnam National University Hanoi.

Christopher Justice received the B.S. and

PhD degree in geography from the
University of Reading, UK in 1973 and
1977 respectively.
He has held a number of different research
positions over the years. From 1983-1997
he served as a visiting scientist at NASA
Goddard Space Flight Center. He has also
served as a research professor at the
University of Virginia, and Research Fellow at the European
Space Agency. Since 2001 he has been a full professor at the
University of Maryland in the Department of Geographical
Sciences and more recently has served as department chair
since 2010. He is the NASA LCLUC program scientist,
member of MODIS science team, global implementation lead
for the GEO Global Agricultural Monitoring Task. He is also
co-director for the center of Global Agricultural Monitoring and
Research. His current research is on land cover and land use
change, global fire, and global agricultural monitoring systems.

Copyright 2017 IEEE. Author submitted final version.

TABLES

Copyright 2017 IEEE. Author submitted final version.

 Table I
SAR incidence angle variation (degrees)

Avg Angle Std Angle Avg angle Std Angle Avg angle Std Angle
Band Date (Double Rice) (Double Rice) (Single Rice) (Single Rice) (Not Rice) (Not Rice)
1 02/03/2016 38.630 3.343 39.304 3.328 39.547 6.414
2 02/15/2016 38.639 3.347 39.314 3.326 39.547 6.414
3 02/27/2016 38.640 3.349 39.316 3.327 39.543 6.414
4 03/10/2016 38.629 3.341 39.306 3.328 39.550 6.412
5 03/22/2016 38.640 3.350 39.326 3.322 39.539 6.411
6 04/03/2016 38.644 3.351 39.334 3.326 39.538 6.409
7 04/15/2016 38.640 3.343 39.325 3.322 39.545 6.411
8 04/27/2016 38.644 3.345 39.327 3.322 39.546 6.411
9 05/09/2016 38.637 3.344 39.322 3.325 39.543 6.410
10 05/21/2016 38.644 3.353 39.331 3.337 39.542 6.408
11 06/02/2016 38.624 3.337 39.306 3.327 39.547 6.412
12 06/14/2016 38.634 3.342 39.315 3.325 39.546 6.412
13 07/08/2016 38.632 3.347 39.303 3.327 39.547 6.410
14 07/20/2016 38.632 3.340 39.313 3.326 39.548 6.412
15 08/01/2016 38.620 3.336 39.302 3.327 39.547 6.412
16 08/13/2016 38.627 3.339 39.303 3.327 39.550 6.411
17 08/25/2016 38.627 3.339 39.304 3.327 39.550 6.412
18 09/06/2016 38.626 3.336 39.307 3.327 39.553 6.412
19 09/18/2016 38.629 3.340 39.307 3.327 39.548 6.412
20 09/30/2016 38.629 3.340 39.310 3.325 39.543 6.412
21 10/12/2016 38.638 3.346 39.325 3.327 39.542 6.410
22 10/24/2016 38.640 3.344 39.325 3.322 39.543 6.411

Local incidence angle, θ, variation in SAR imagery per mapped land cover class for VV&VH 10m
dataset: Minimal variation through time in all categories. ‘Not rice’ areas have higher standard
deviation due to mountainous areas.

Copyright 2017 IEEE. Author submitted final version.

 Table II
Dataset descriptions
Dataset
Description
name

VV10m Vertical-vertical (VV) polarized bands at 10m spatial resolution

VH10m Vertical-horizontal (VH) polarized bands at 10m spatial resolution

VVVH10m Both VV and VH bands at 10m resolution

VV20m VV bands at 20m spatial resolution

VH20m VH bands at 20m spatial resolution

VVVH20m Both VV and VH bands at 20m spatial resolution

Copyright 2017 IEEE. Author submitted final version.

 REFERENCE REFERENCE
VV and VH Class Double rice Single rice Non-rice Total Adj. User's Acc. Area (ha) VV and
VH Class Double rice Single rice Non-rice Total Adj. User's Acc. Mapped Area (ha)
Double rice 420 2 17 439 95.7% (0.09%) 214,565
Double rice 120 1 6 127 94.5% (0.35%) 220,356
Single rice 0 160 1 161 99.4% (0.10%) 4,719 Single rice 1 48 1 50 95.2% (1.02%) 5,533
Non-rice 68 4 916 988 92.7% (0.05%) 588,449 Non-rice 24 3 299 326 89.5% (0.18%) 581,839
Total 488 166 934 1588 807,733 Total 145 52 306 503 - 807,729
Adj. Producer's 83.5% 98.5% 98.1%
Acc. (0.15%) 58.3% (0.59%) (0.03%) Adj. Producer's Acc. 82.9% (0.51%) 42.8% (1.90%) (0.09%)

Overall Accuracy: 93.5% (1.33%)

Overall Accuracy: 92.5% (2.52%)
REFERENCE REFERENCE
VH Class Double rice Single rice Non-rice Total Adj. User's Acc. (C.I) Area (ha) Adj. User's
VH Class Double rice Single rice Non-rice Total Acc. Mapped Area (ha)
Double rice 410 0 13 423 96.9% (0.08%) 212,465
Double rice 117 0 7 124 94.4% (0.37%) 218,788
Single rice 0 161 3 164 98.2% (0.16%) 4,319
Single rice 0 48 1 49 98.0% (0.58%) 5,227
Non-rice 78 5 918 1001 91.7% (0.05%) 590,950 Non-rice 28 4 298 330 90.3% (0.18%) 583,713
Total 488 166 934 1588 807,733 Total 145 52 306 503 - 807,729
Adj. Producer's 81.7% 98.8% 97.7%
Acc (0.16%) 59.0% (0.58%) (0.02%) Adj. Producer's Acc. 80.7% (0.54%) 42.0% (1.90%) (0.10%)

Overall Accuracy: 93.1% (1.38%) Overall Accuracy: 91.4% (2.68%)

REFERENCE REFERENCE
VV Class Double rice Single rice Non-rice Total Adj. User's Acc. (C.I) Area (ha) Adj. User's
VV Class Double rice Single rice Non-rice Total Acc. Mapped Area (ha)
Double rice 389 3 14 406 95.8% (0.10%) 208,276
Double rice 116 2 8 126 92.1% (0.42%) 214,903
Single rice 0 145 1 146 99.3% (0.11%) 4,908
Single rice 0 48 0 48 100% (0.00%) 5,450
Non-rice 99 13 919 1031 89.1% (0.06%) 594,549
Non-rice 29 2 298 329 90.6% (0.17%) 587,376
Total 488 161 934 1583 807,733 Total 145 52 306 503 807,729
Adj. Producer's 77.8% 98.7% 97.5%
Acc (0.17%) 35.0% (0.58%) (0.02%) Adj. Producer's Acc. 79.3% (0.55%) 43.8% (1.91%) (0.10%)

Overall Accuracy: 90.9% (1.57%) Overall Accuracy: 91.0% (2.75%)

confusion matrices with associated pixels per class, mapped class area, and adjusted accuracies with confidence intervals based on calculated unbiased areal
estimates for a) 10m data; and b) 20m data.

Copyright 2017 IEEE. Author submitted final version.

Figures

Copyright 2017 IEEE. Author submitted final version.

 Figure 1
Time-series SAR signal

Time-series SAR signal over different land cover types and polarizations in the study area.

Copyright 2017 IEEE. Author submitted final version.

 Figure 2
Multitemporal SAR composite

Multi-temporal SAR composite with training data overlaid. Note: border

thickness of training data exaggerated for visualization

Copyright 2017 IEEE. Author submitted final version.

 Figure 3
Landscape analysis flowchart

Landscape-scale analysis using Fragstats and mean-shift image segmentation were

used to derive: 1) continuous rice tracts/patches, 2) size and number of rice
collectives; 3) the total number of paddy rice fields
Copyright 2017 IEEE. Author submitted final version.
 Figure 4
Paddy rice maps

Figure 3 includes each SAR rice map and a pixel count of double-rice agreement between the 6 datasets on the right side.
Copyright 2017 IEEE. Author submitted final version.
 5

Area (ha)

Copyright 2017 IEEE. Author submitted final version.

 Figure 6
Rice Areal Variation

Rice areal variation zoomed to

Hanoi province highlighting
Copyright 2017 IEEE. Author submitted final version.
variation in each dataset.
 Figure 7
Province-level rice variation
Double Rice Single Rice

Z-score

Z-score
Double rice and single cropped rice province-level z-scores indicating deviation from the mean.

Copyright 2017 IEEE. Author submitted final version.

 Figure 8
Commune-level rice variation

Average absolute deviation and coefficient of variation for double rice area per commune with province boundaries overlaid. Results
suggest most AAD in communes with high rice mapped areas, while CV results suggest communes in the north tended to have
higher percentage variation likely attributed to mapped area difficulties in more mosaicked landscapes with smaller fields.
Copyright 2017 IEEE. Author submitted final version.
 Figure 9
Commune-level rice phenology

Commune-level sowing and harvest day of year (DoY) ranges based on Sentinel-1 signal. Reason for
Copyright 2017 IEEE. Author submitted final version.
minimal variation in summer sowing is due to missing data for ~06/28/2016, thus actual range is bigger.
 Figure 10
Mean Decrease in Accuracy

Mean decrease in out of bag accuracy (%): suggests bands from all parts of rice stage
(planting/growth/harvest) are important, with bands from second season planting and first season harvest
Copyright
generally most critical. Note for VVVH 2017 IEEE.
graphs, Author submitted
bands 1-22 are final version.
VV and bands 23-44 are VH.